Data reorganization in parallel database systems

Abstract

Parallel database systems are suitable for use in applications with high capacity and high performance and availability requirements. The trend in such systems is to provide efficient online capability for performing various system administration functions such as, index creation and maintenance, backup/restore, reorganization, and gathering of statistics. For some of these functions the online capability can be efficiently supported by the use of incremental algorithms, i.e., algorithms that achieve the function in several, relatively small (i.e., less time-consuming) steps, rather than in a single, large step. Incremental algorithms ensure that only small parts of the database become inaccessible for short durations as opposed to nonincremental algorithms which may lock large portions of the database or the entire database for a longer duration. The authors discuss issues in providing concurrent data reorganization capability using incremental algorithms in parallel database systems.